Diabetes and obesity in heart failure: a crossroads for prognosis and the efficacy of beta-blockers

//Presented at ESC Heart Failure 2018 in Vienna, Austria, by Dipak Kotecha (Birmingham, UK) //

Introduction and methods

It is well-known that uptake of guideline-recommended therapy is suboptimal, and diabetes in heart failure (HF) is associated with increased mortality. And, while not everybody agrees on why it happens, there is little discussion about the existence of an obesity paradox in HF. The prognostic impact of having both diabetes and obesity in HF is unknown to date. Moreover, it remains to be determined whether beta-blockers are effective at reducing mortality in these patients with HF with reduced ejection fraction (HFrEF), in sinus rhythm.

That is why the Beta-blockers in HF Collaborative Group set out to pool individual patient data from 18,637 HF patients in double-blind randomized controlled trials (RCTs) that reported mortality as a major trial endpoint. Patients with arrythmia, including atrial fibrillation, were excluded, as were those without a diabetes status at baseline, and those with missing baseline BMI. RCT evidence is considered the ‘gold-standard’, but this is only true when strict standards are followed, as in this study. 8298 HF patients without diabetes and obesity were included in this analysis, 2331 non-obese diabetes patients, 2121 obese patients without diabetes and 1109 obese diabetes patients. Mean follow-up in the 11 included studies was 1.4 years.

Main results

• With the no diabetes, non-obese patients as a reference (annualized mortality: 10.3%), the adjusted mortality risk was increased for those who are non-obese but have diabetes (15.7%), while is was lower in obese patients without diabetes (7.3%) and the same in patients with diabetes and obesity (10.4%). Thus, the two factors contradict each other.
• In patients without diabetes, beta-blockers significantly lowered all-cause mortality (HR: 0.68, 95%CI: 0.61-0.75, P<0.001) and CV death (HR: 0.67, 95%CI: 0.59-0.75, P<0.001).
• In patients with diabetes, beta-blocker treatment significantly lowered all-cause mortality (HR: 0.81, 95%CI: 0.69-0.94, P=0.008) and CV mortality (HR: 0.80, 95%CI: 0.68-0.95, P=0.010).
• For all-cause mortality, diabetes status showed a significant interaction with the treatment effect (P=0.049), and a trend for an interaction was seen with CV mortality (P=0.063).
• In patients without obesity, beta-blockers significantly lowered all-cause mortality (HR: 0.69, 95%CI: 0.62-0.76, P<0.001) and CV death (HR: 0.68, 95%CI: 0.61-0.75, P<0.001)(P-interaction: 0.04).
• In patients with obesity, beta-blocker treatment did not significantly lower all-cause mortality (HR: 0.84, 95%CI: 0.69-1.03, P=0.092), nor CV mortality (HR: 0.84, 95%CI: 0.68-1.04, =0.11)(P-interaction: 0.051).

Conclusions

These large individual datasets show that beta-blocker treatment lower the risk for all-cause mortality in both diabetics and non-diabetics without obesity. This is a clear impetus to prescribe guideline-recommended therapy. Diabetes HF patients who are not obese, have the highest mortality rate (in sinus rhythm). In those with both diabetes and obesity, the benefit is less certain, as obesity attenuates the prognostic impact of beta-blockers, especially in diabetics.

The large sample size with double-blind RCT data and individual patient data is certainly a strength of this study. Despite the large data set, data for BMI <18.5 and >50 was scarce. Another limitation is that diabetes is not a yes/no health condition; no data on HbA1c or blood glucose levels was available and used. Moreover, the diabetic therapy patients may have taken was not known.

Discussion

The discussant noted that data on the obesity paradox is scarce. This study adds to some existing clinical proof that there is a different obesity paradox in patients that have diabetes. What causes this, is topic of speculation. Possibly the volume of the drug administered to patients may have played a role in other studies, but in this study the dose was increased in obese patients. Other speculated mechanisms include autonomic activation, but remains to be clarified.

Our reporting is based on the information provided at the ESC Heart Failure 2018 congress